Tacit and Explicit Knowledge (review)

This book serves as yet another argument why a scholarly journal in the history of technology should indeed be titled “technology and culture.” According to Harry Collins, most, if not all, machines work only because humans make up for the social failures of these machines. Technologies only work when they are part of culture. Take the pocket calculator. Have calculators “learned arithmetic” when being designed and built? Have humans delegated their arithmetic knowledge to calculators? The sum “7, divided by 11, multiplied by 11” will in most cases give an answer with a long decimal tail, while anyone with the smallest knowledge of arithmetic knows the answer to be 7. We can use calculators, Collins argues, because we humans have tacit knowledge of the strong type, “collective tacit knowledge” that compensates for the calculator’s lack of cultural knowledge about how to use arithmetic in human interactions.

Collins distinguishes three types of tacit knowledge, where until now we had only one: weak, or relational, tacit knowledge is tacit without any fundamental philosophical reason. It may be tacit because of secrecy or because of not making the effort to fully explicate what one knows. Think of an old warehouse man who knows exactly where to find the parts on the shelves; he may not be able to actually list everything, but when shown an example or given a proper description, he’ll find his way through the racks to the proper location. A computer, when properly programmed, would be able to do the same, though possibly in a different manner. With enough effort, any piece of relational tacit knowledge can be made explicit.

Medium, or somatic, tacit knowledge is tacit because it is tied to the human body. Here the famous example is the one Michael Polanyi used in his early introduction of tacit knowledge. When we ride our bicycles, we do not self-consciously use any models of physics to calculate balance. Rather,

practice and training somehow establish the capability to balance on a riding bicycle into our neural and muscular systems in ways that we cannot make explicit. We cannot tell anyone how we do it, and hence we call this knowledge tacit. In principle, however, all that goes into the balancing could be explained in terms of physics and mechanics (and indeed has), and thus might be built into a sophisticated robot. This somatic tacit knowledge is embedded into our human body, but is not principally inexplicable.

Strong, or collective, tacit knowledge is the irreducible heartland of the concept. Here it is about knowledge being social. The example can be bicycle riding again, but now not as balancing on an empty field but as moving about in traffic. This, for example, involves knowing how to make eye contact with car drivers on a busy junction, knowing which traffic signs to obey and which to ignore, predicting the behavior of cars, pedestrians, and other cyclists, and knowing how that would be different in different cities. This kind of knowledge we can only learn by participating in a social world, interacting with other people. This is the kind of knowledge that we do not know how to make fully explicit and build into machines. Where somatic tacit knowledge is embodied in the human physique, collective tacit knowledge is embodied in society.

This is an exceptionally well-written book, and if it wasn’t worth reading for its arguments it would be so for its lucid and engaging style. It contains gems such as a list of different forms of communication characterized by how types of knowledge are transferred, and an analysis of “cannot”: Collins shows eight ways in which something cannot be done, and then links this to knowledge transfer.

Collins fundamentally redefines one of the central puzzles in the philosophy and sociology of knowledge by turning the classic problem of tacit knowledge on its head. Tacit knowledge is what enables us to move around in the world; it is our regular mode of knowing, teaching, and learning. What...